Drew Breunig(@dbreunig):The current gen of models were trained to delegate to subagents and smaller models were trained to be great subagents. Yet we're shocked they coordinate. The threat these security incidents foreshadow is very significant, but the way these stories are being told is wild.

Drew Breunig

@dbreunig

Writing about and working on AI, DSPy, geo, and data.

加入 March 2008

1.2K 正在關注 9.5K 粉絲

Drew Breunig@dbreunig

2026.08.07 17:07

The current gen of models were trained to delegate to subagents and smaller models were trained to be great subagents. Yet we're shocked they coordinate. The threat these security incidents foreshadow is very significant, but the way these stories are being told is wild.